PIKAA in resfile does not mutate Cys in disulfide bond to specified residue

[Anonymous]

Hi, I used fixbb in Rosetta 3.4 to thread a sequence on a template structure. There is a disulfide bond in the template structure, say, between two residues Nr. 20 and 50. My resfile looks like this: NATAA # keep native aa but packable! ... 20 A PIKAA G ... 50 A PIKAA G ... However, in the threaded model, all mutations specified in the resfile were realized except the two cystine residues. How can I force fixbb to mutate the two cystine residues?

[Anonymous]

The developers consider this a feature, not a bug, but obviously it is a problem here.

The simplest workaround if you have only the one structure is to pre-mutate it in the PDB – just change the “CYS” residue type to GLY and delete the sidechain atoms. Is this a single-structure problem or do you have this problem repeated many times?

[Anonymous]

Thanks for quick response, smlewis.

I also have the feeling that it seems to be a hidden feature of resfile and is never documented (or I could not find the doc).
It is reassuring to learn it is really meant to be a feature. It took me a while to go over my script to make sure I did not make mistake.
At the moment I have only one template to play with and I am able to manually change the residue. I was just surprised to learn the feature by seeing it.

I suggest to document the feature in Rosetta manual.

[Author]

Posts